Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Modeling IoT Traffic Patterns: Insights from a Statistical Analysis of an MTC Dataset

David E. Ruiz-Guirola, Onel L. A. Løpez, Samuel Montejo-Sanchez

TL;DR

This paper analyzes the Smart Campus MTC dataset to model IoT traffic via two targeted stochastic models: a Poisson point process (PPP) for event-driven traffic and a quasi-periodic model for periodic updating traffic. It uses goodness-of-fit tests—including Kolmogorov-Smirnov, Anderson-Darling, and chi-squared tests—and RMSE to compare models against real data, identifying PPP as the best fit for ED patterns and the quasi-periodic model for PU patterns, with tail errors below $0.12$ and $0.08$ respectively. The study formalizes a system model with 2D PPP deployments of MTDs and event epicenters, event-generation mechanisms, and a bursty payload-exchange process captured by a geometric distribution, while validating the approaches on 11 months of real data. These results provide practical, data-driven templates for realistic MTC traffic simulation, prediction, and energy-efficient management in IoT networks.

Abstract

The Internet-of-Things (IoT) is rapidly expanding, connecting numerous devices and becoming integral to our daily lives. As this occurs, ensuring efficient traffic management becomes crucial. Effective IoT traffic management requires modeling and predicting intrincate machine-type communication (MTC) dynamics, for which machine-learning (ML) techniques are certainly appealing. However, obtaining comprehensive and high-quality datasets, along with accessible platforms for reproducing ML-based predictions, continues to impede the research progress. In this paper, we aim to fill this gap by characterizing the Smart Campus MTC dataset provided by the University of Oulu. Specifically, we perform a comprehensive statistical analysis of the MTC traffic utilizing goodness-of-fit tests, including well-established tests such as Kolmogorov-Smirnov, Anderson-Darling, chi-squared, and root mean square error. The analysis centers on examining and evaluating three models that accurately represent the two most significant MTC traffic types: periodic updating and event-driven, which are also identified from the dataset. The results demonstrate that the models accurately characterize the traffic patterns. The Poisson point process model exhibits the best fit for event-driven patterns with errors below 11%, while the quasi-periodic model fits accurately the periodic updating traffic with errors below 7%.

Modeling IoT Traffic Patterns: Insights from a Statistical Analysis of an MTC Dataset

TL;DR

This paper analyzes the Smart Campus MTC dataset to model IoT traffic via two targeted stochastic models: a Poisson point process (PPP) for event-driven traffic and a quasi-periodic model for periodic updating traffic. It uses goodness-of-fit tests—including Kolmogorov-Smirnov, Anderson-Darling, and chi-squared tests—and RMSE to compare models against real data, identifying PPP as the best fit for ED patterns and the quasi-periodic model for PU patterns, with tail errors below and respectively. The study formalizes a system model with 2D PPP deployments of MTDs and event epicenters, event-generation mechanisms, and a bursty payload-exchange process captured by a geometric distribution, while validating the approaches on 11 months of real data. These results provide practical, data-driven templates for realistic MTC traffic simulation, prediction, and energy-efficient management in IoT networks.

Abstract

The Internet-of-Things (IoT) is rapidly expanding, connecting numerous devices and becoming integral to our daily lives. As this occurs, ensuring efficient traffic management becomes crucial. Effective IoT traffic management requires modeling and predicting intrincate machine-type communication (MTC) dynamics, for which machine-learning (ML) techniques are certainly appealing. However, obtaining comprehensive and high-quality datasets, along with accessible platforms for reproducing ML-based predictions, continues to impede the research progress. In this paper, we aim to fill this gap by characterizing the Smart Campus MTC dataset provided by the University of Oulu. Specifically, we perform a comprehensive statistical analysis of the MTC traffic utilizing goodness-of-fit tests, including well-established tests such as Kolmogorov-Smirnov, Anderson-Darling, chi-squared, and root mean square error. The analysis centers on examining and evaluating three models that accurately represent the two most significant MTC traffic types: periodic updating and event-driven, which are also identified from the dataset. The results demonstrate that the models accurately characterize the traffic patterns. The Poisson point process model exhibits the best fit for event-driven patterns with errors below 11%, while the quasi-periodic model fits accurately the periodic updating traffic with errors below 7%.
Paper Structure (21 sections, 6 equations, 9 figures, 1 table)

This paper contains 21 sections, 6 equations, 9 figures, 1 table.

Table of Contents

  1. Introduction
  2. Dataset and Scenario
  3. Statistical analysis and Traffic models
  4. System model
  5. Influence of an Event Epicenter
  6. Generation of Events
  7. Poisson model
  8. Quasi-periodic traffic
  9. Payload exchange
  10. Performance metrics
  11. RMSE
  12. Kolmogorov-Smirnov test
  13. Anderson-Darling test
  14. Chi-squared test
  15. Analysis Results
  16. ...and 6 more sections

Figures (9)

  • Figure 1: Map of the University of Oulu Smart Campus.
  • Figure 2: Types of sensors and their readings.
  • Figure 3: Illustration of an MTC network where a BS controls and collects information from the MTDs.
  • Figure 4: Traffic exchanged between MTDs and the BS is modeled as a two-state complete Markov chain.
  • Figure 5: Extract of a) Poisson (top) and b) quasi-periodic (bottom) traffic. The alarms represent the time slots (x-axis) where a given MTD (y-axis) is active.
  • ...and 4 more figures